Stochastic Variational Deep Kernel Learning

TL;DR

This paper introduces a novel deep kernel learning model that combines deep neural networks with Gaussian processes, enabling scalable classification and multi-task learning with improved performance on large datasets.

Contribution

It proposes a new stochastic variational inference method for deep kernel learning that handles additive covariance structures and large-scale data efficiently.

Findings

Outperforms standalone deep networks, SVMs, and Gaussian processes on benchmarks.

Achieves scalable training on datasets with millions of points.

Demonstrates effectiveness on airline delay, CIFAR, and ImageNet datasets.

Abstract

Deep kernel learning combines the non-parametric flexibility of kernel methods with the inductive biases of deep learning architectures. We propose a novel deep kernel learning model and stochastic variational inference procedure which generalizes deep kernel learning approaches to enable classification, multi-task learning, additive covariance structures, and stochastic gradient training. Specifically, we apply additive base kernels to subsets of output features from deep neural architectures, and jointly learn the parameters of the base kernels and deep network through a Gaussian process marginal likelihood objective. Within this framework, we derive an efficient form of stochastic variational inference which leverages local kernel interpolation, inducing points, and structure exploiting algebra. We show improved performance over stand alone deep networks, SVMs, and state of the art scalable Gaussian processes on several classification benchmarks, including an airline delay dataset containing 6 million training points, CIFAR, and ImageNet.